v4.44.2

fix save_pretrained

setup.py

@@ -430,7 +430,7 @@ install_requires = [
 
 setup(
     name="transformers",
-    version="4.44.1",  # expected format is one of x.y.z.dev0, or x.y.z.rc1 or x.y.z (no to dashes, yes to dots)
+    version="4.44.2",  # expected format is one of x.y.z.dev0, or x.y.z.rc1 or x.y.z (no to dashes, yes to dots)
     author="The Hugging Face team (past and future) with the help of all our contributors (https://github.com/huggingface/transformers/graphs/contributors)",
     author_email="transformers@huggingface.co",
     description="State-of-the-art Machine Learning for JAX, PyTorch and TensorFlow",

src/transformers/__init__.py

@@ -18,7 +18,7 @@
 # to defer the actual importing for when the objects are requested. This way `import transformers` provides the names
 # in the namespace without actually importing anything (and especially none of the backends).
 
-__version__ = "4.44.1"
+__version__ = "4.44.2"
 
 from typing import TYPE_CHECKING
 

src/transformers/models/jamba/modeling_jamba.py

@@ -210,6 +210,7 @@ class HybridMambaAttentionDynamicCache(DynamicCache):
     """
 
     def __init__(self, config, batch_size, dtype=torch.float16, device=None):
+        super().__init__()
         self.dtype = dtype
         self.layers_block_type = config.layers_block_type
         self.has_previous_state = False  # only used by mamba
@@ -629,7 +630,7 @@ class JambaMambaMixer(nn.Module):
 
         # S4D real initialization. These are not discretized!
         # The core is to load them, compute the discrete states, then write the updated state. Keeps the memory bounded
-        A = torch.arange(1, self.ssm_state_size + 1, dtype=torch.float32)[None, :]
+        A = torch.arange(1, self.ssm_state_size + 1)[None, :]
         A = A.expand(self.intermediate_size, -1).contiguous()
 
         self.A_log = nn.Parameter(torch.log(A))

src/transformers/processing_utils.py

@@ -522,7 +522,7 @@ class ProcessorMixin(PushToHubMixin):
                 token=kwargs.get("token"),
             )
 
-        if set(self.to_dict().keys()) == {"processor_class"}:
+        if set(processor_dict.keys()) == {"processor_class"}:
             return []
         return [output_processor_file]